probabilistic neural network error

i have a cource code like this, and i have error, any body help me ... plisss :(

it's a part of gambar1.m

mm=10;
d=imread(filename)
r=d;
m=fspecial('unsharp');
x=imfilter(r,m);

      data=x;
     pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\Hish_Pass_Filter\';
    % mkdir(pathName);
  imwrite(data,[pathname, 'HPV','.jpg']);
  [r,c]=size(x);
  max=0;
  min=0;
  for j=2:r-1
      for i=2:c-1
          p=i-1;
          q=i+1;
          a=j-1;
          b=j+1;
          d1=abs(x(j,i)-x(j,p));
          d2=x(j,i)-x(j,q);
          e1=x(j,i)-x(a,i);
          e2=x(j,i)-x(b,i);
          f1=x(j,i)-x(a,p);
          f2=x(j,i)-x(b,q);
          g1=x(j,i)-x(a,q);
          g2=x(j,i)-x(b,p);
          if (d1 < 0)
              d1=d1*(-1);
          end
          if (d2 < 0)
              d2=d2*(-1);
          end
          if (e1 < 0)
              e1=e1*(-1);
          end
          if (e2 < 0)
              e2=e2*(-1);
          end
            if (f1 < 0)
                f1=f1*(-1);
            end
            if (f2 < 0)
                f2=f2*(-1);
            end
            if (g1 < 0)
                g1=g1*(-1);
            end
            if (g2 < 0)
                g2=g2*(-1);
            end
            if (d1 >= mm) && (d2 >= mm)
                x(j,i)=255;
            elseif (e1 >= mm) && (e2 >= mm)
                x(j,i)=255;
            elseif (f1 >= mm) && (f2 >= mm)
                x(j,i)=255;
            elseif (g1 >= mm) && (g2 >= mm)
                x(j,i)=255;
            else
                x(j,i)=0;  
            end
        end
    end

      e=x

     data=e;
     pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\Gambar_Canny\';
    % mkdir(pathName);
  imwrite(data,[pathname, 'Canny','.jpg']);
  sakita=reshape(x,[10000 1]);

data=sakita;
   pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\';
  % mkdir(pathName);
imwrite(data,[pathname, 'sakita','.jpg']);

function pushbutton2_Callback(hObject, eventdata, handles)
% hObject    handle to pushbutton2 (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)

%function KikyPNN()

%ini adalah contoh kode program membuat PNN menggunakan fungsi newpnn
%data yang akan di testing disimpan dalam matriks Testdata
%hasil testing akan disimpan dalam matriks Yc.
%bandingkan hasiltestdata dengan yang seharusnya

%gunakan perintah “help newpnn” pada Command Line untuk mempelajari menggunakan fungsi newpnn
%clear;
%for handles.aa=handles.1:1:handles.30
 %   handles.aa=[handles.aa:hanldes.aa]
%end

% DIR returns as a structure array.  You will need to use () and . to get
% the file names.
A = gambar1(hObject, eventdata, handles);
aa=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakita.jpg');
B = gambar2(hObject, eventdata, handles);
bb=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakitb.jpg');
C = gambar3(hObject, eventdata, handles);
cc=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakitc.jpg');
D = gambar4(hObject, eventdata, handles);
dd=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakitd.jpg');
E = gambar5(hObject, eventdata, handles);
ee=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakite.jpg');
F = gambar6(hObject, eventdata, handles);
ff=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakitf.jpg');
G = gambar7(hObject, eventdata, handles);
gg=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakitg.jpg');
H = gambar8(hObject, eventdata, handles);
hh=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakith.jpg');
I = gambar9(hObject, eventdata, handles);
ii=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakiti.jpg');
J = gambar10(hObject, eventdata, handles);
jj=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakitj.jpg');
K = gambar11(hObject, eventdata, handles);
kk=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehata.jpg');
L = gambar12(hObject, eventdata, handles);
ll=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehatb.jpg');
M = gambar13(hObject, eventdata, handles);
mm=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehatc.jpg');
N = gambar14(hObject, eventdata, handles);
nn=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehatd.jpg');
O = gambar15(hObject, eventdata, handles);
oo=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehate.jpg');
P = gambar16(hObject, eventdata, handles);
pp=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehatf.jpg');
Q = gambar17(hObject, eventdata, handles);
qq=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehatg.jpg');
R = gambar18(hObject, eventdata, handles);
rr=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehath.jpg');
S = gambar19(hObject, eventdata, handles);
ss=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehati.jpg');
T = gambar20(hObject, eventdata, handles);
tt=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sehatj.jpg');

%Data(1:10000,1:3)=zeros;

%Data(1,1:10000)=sakita';
%Data1=sakita';
%Data(2,:)=sakitb;

%Data(3,:)=sehata;
%tic

%Dataasli=Data
%a=ones(1,10000);
%b=2*ones(1,10000);
%clear handles.aa;
%Dataasli = [gambar1 gambar2 gambar3] % terjadi pelupingan data
Dataasli = [aa bb cc dd ee ff gg hh ii jj kk ll mm nn oo pp qq rr ss tt];
%Dataasli=[1 1 2]
Kelasdataasli = [1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2];
T = ind2vec(Kelasdataasli)
net = newpnn(Dataasli,T)
%mm=input('Input your threshold between 0 and 255    ');
testdata=[handles.f]
% bila test data sudah benar, ,maka hasil dari test data yang sesuai dengan
% data asli masuk otomatis kedalam data asli ututan +1
%testdata=[handles.g]
%testdata=Y=reshape(x,1)
Y = sim(net,testdata)
Yc = ind2vec(Y)
hasiltestdata=Yc

an the error is :

T =

     (1,1)        1
     (1,2)        1
     (1,3)        1
     (1,4)        1
     (1,5)        1
     (1,6)        1
     (1,7)        1
     (1,8)        1
     (1,9)        1
     (1,10)       1
     (2,11)       1
     (2,12)       1
     (2,13)       1
     (2,14)       1
     (2,15)       1
     (2,16)       1
     (2,17)       1
     (2,18)       1
     (2,19)       1
     (2,20)       1

Error using network/subsasgn>network_subsasgn (line 536)
net.IW{1,1} must be a 20-by-10000 matrix.

Error in network/subsasgn (line 13)
net = network_subsasgn(net,subscripts,v,netname);

Error in newpnn>create_network (line 122)
  net.iw{1,1} = p';

Error in newpnn (line 66)
    net = create_network(param);

Error in aplikasi>pushbutton2_Callback (line 301)
net = newpnn(Dataasli,T)

Error in gui_mainfcn (line 96)
        feval(varargin{:});

Error in aplikasi (line 42)
    gui_mainfcn(gui_State, varargin{:});

Error in
@(hObject,eventdata)aplikasi('pushbutton2_Callback',hObject,eventdata,guidata(hObject))

Error while evaluating uicontrol Callback

Answer 1

[Walter Roberson]

JPEG images cannot be pseudocolor (2 dimensional) images, so each imread() you are doing is reading in a truecolor (3 dimensional) array. You horzcat() the 20 images together, which is going to still result in a 3 dimensional array. You pass that to newpnn() as the first argument. newpnn() is only documented as accepting 2D arrays for its first argument.

[deniel munthe]

would you want give me example with horzcat() , i really nead it.

[deniel munthe]

sir can you help me. oh, before fspacial, images have canges to grayscale before. im use imread, for inisialisation images before it

[Walter Roberson]

The [ ] list operator is already horzcat(), except when it sees semi-colon

Your line

Dataasli = [aa bb cc dd ee ff gg hh ii jj kk ll mm nn oo pp qq rr ss tt];

is already the same as

Dataasil = horzcat(aa,bb,cc,dd,ee,ff,gg,hh,ii,jj,kk,ll,mm,nn,oo,pp,qq,rr,ss,tt];

However, since

aa=imread('D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\sakita.jpg');

aa is going to be a 3D array, and the other variables are going to be 3D arrays (red, green, blue for each pixel), so when you horzcat() them together you get a 3D array as a result.

If you do not want a 3D array as a result, then you probably should not be using [ ] on 3D arrays.

[deniel munthe]

before i'm use imread, i have a process for images sakitaa before. n source code is

function [sakita]=gambar1(hObject, eventdata, handles)

% hObject handle to pushbutton1 (see GCBO)

% eventdata reserved - to be defined in a future version of MATLAB

% handles structure with handles and user data (see GUIDATA)

a=imread('D:\Deniel Data\SKRIPSI\APLIKASI\sel sakit\sakita.jpg');

b=a;

c=rgb2gray(b);

d=imresize(c,[100 100]);

data=d;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\Gambar_Grayscale\';

% mkdir(pathName);

imwrite(data,[pathname, 'grayscale','.jpg']);

mm=10;

r=d;

m=fspecial('unsharp');

x=imfilter(r,m);

data=x;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\Hish_Pass_Filter\';

% mkdir(pathName);

imwrite(data,[pathname, 'HPV','.jpg']);

[r,c]=size(x);

max=0;

min=0;

for j=2:r-1

for i=2:c-1

p=i-1;

q=i+1;

a=j-1;

b=j+1;

d1=abs(x(j,i)-x(j,p));

d2=x(j,i)-x(j,q);

e1=x(j,i)-x(a,i);

e2=x(j,i)-x(b,i);

f1=x(j,i)-x(a,p);

f2=x(j,i)-x(b,q);

g1=x(j,i)-x(a,q);

g2=x(j,i)-x(b,p);

if (d1 < 0)

d1=d1*(-1);

end

if (d2 < 0)

d2=d2*(-1);

end

if (e1 < 0)

e1=e1*(-1);

end

if (e2 < 0)

e2=e2*(-1);

end

if (f1 < 0)

f1=f1*(-1);

end

if (f2 < 0)

f2=f2*(-1);

end

if (g1 < 0)

g1=g1*(-1);

end

if (g2 < 0)

g2=g2*(-1);

end

if (d1 >= mm) && (d2 >= mm)

x(j,i)=255;

elseif (e1 >= mm) && (e2 >= mm)

x(j,i)=255;

elseif (f1 >= mm) && (f2 >= mm)

x(j,i)=255;

elseif (g1 >= mm) && (g2 >= mm)

x(j,i)=255;

else

x(j,i)=0;

end

end

end

e=x

data=e;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\Gambar_Canny\';

% mkdir(pathName);

imwrite(data,[pathname, 'Canny','.jpg']);

sakita=reshape(x,[10000 1])

data=sakita;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\';

% mkdir(pathName);

imwrite(data,[pathname, 'sakita','.bmp']);

%sakita=reshape(data,[10000 1]);

[Walter Roberson]

You write out a number of images, but you are not writing out the images that you imread() in pushbutton2_Callback

[deniel munthe]

so what i must to do for fix it, i'm very stack here.

[deniel munthe]

owh i'm sorry before,

it's the right code when i'm use for pushbutton-callback2.

end

end

end

e=x

data=e;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\Gambar_Canny\';

% mkdir(pathName);

imwrite(data,[pathname, 'Canny','.jpg']);

sakita=reshape(x,[10000 1])

data=sakita;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\';

% mkdir(pathName);

imwrite(data,[pathname, 'sakita','.jpg']);

%sakita=reshape(data,[10000 1]);

end

end

end

e=x

data=e;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\Gambar_Canny\';

% mkdir(pathName);

imwrite(data,[pathname, 'Canny','.jpg']);

sakita=reshape(x,[10000 1])

data=sakita;

pathname = 'D:\Deniel Data\SKRIPSI\APLIKASI\ReShape\';

% mkdir(pathName);

imwrite(data,[pathname, 'sakita','.jpg']);

%sakita=reshape(data,[10000 1]);

i'm use .jpg not .bmp, but i'm always have an error

[Walter Roberson]

Okay, to check: what does size(aa) show? And size(Dataasli) and size(T) ?

[deniel munthe]

aa is <10000x1 uint8> min 0 , max 255, Dataasli is <10000x20 uint8> min 0 max 255,

Kelasdataasli = [1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2];

T = ind2vec(Kelasdataasli)

T is <2x20 sparse double>

now, what i must to do ?

[deniel munthe]

are you have any idea ?

[Walter Roberson]

I would suggest _experimenting_ as

newpnn(Dataasli .', T)

[deniel munthe]

have error with net=newpnn(Dataasli .', T)

Error using network/subsasgn>network_subsasgn (line 536)

net.IW{1,1} must be a 20-by-20 matrix.

Error in network/subsasgn (line 13)

net = network_subsasgn(net,subscripts,v,netname);

Error in newpnn>create_network (line 122)

net.iw{1,1} = p';

Error in newpnn (line 66)

net = create_network(param);

Error in aplikasi>pushbutton2_Callback (line 302)

net=newpnn(Dataasli .', T)

Error in gui_mainfcn (line 96)

feval(varargin{:});

Error in aplikasi (line 42)

gui_mainfcn(gui_State, varargin{:});

Error in

@(hObject,eventdata)aplikasi('pushbutton2_Callback',hObject,eventdata,guidata(hObject))

Error while evaluating uicontrol Callback

[Walter Roberson]

Sorry, I do not know. I do not have that toolbox to test with.